On Mon, Jun 15, 2015 at 9:04 AM, YKY (Yan King Yin, 甄景贤) < [email protected]> wrote:
> > > I found out that a "distributive representation" > > does not come with superposition (I don't recall where I got that idea > from). > > For example, 100 neurons which take only binary (0,1) values can represent > maximally 2^100 different "states". This is vastly bigger than the number > of states for a completely local representation, which would be 100. > > But now we have no way to superimpose the states -- all the available bits > are used up. > > I have to research a bit about the idea of superposition... to see how > it gels with distributive representations. > > PS: if we use a dimension higher than the dimension of the signal, that > representation is called "over-complete", and mathematically it's called a > "frame" (the famous example being the "Mercedes Benz" tri-vector "basis" > for 2-D space). > There are ways to use such representations to increase accuracy or combat > noise. > Could you explain this a little more. I found references but I could not really get what you are getting at. For example, are any multiple vectors used as a "basis" then "over-complete" and called a "frame"? What does it mean to use multiple vectors as a basis? How is that useful? Would they be used to increase accuracy because of round off errors (like taking multiple readings to determine the position of a broadcast from a boat) or are you talking about something else? Jim Bromer ------------------------------------------- AGI Archives: https://www.listbox.com/member/archive/303/=now RSS Feed: https://www.listbox.com/member/archive/rss/303/21088071-f452e424 Modify Your Subscription: https://www.listbox.com/member/?member_id=21088071&id_secret=21088071-58d57657 Powered by Listbox: http://www.listbox.com
